9 papers · 1 filter
Scaling Laws for Looped Mixture of Experts
Yanbei Chen, Anirudh Goyal, Raghuraman Krishnamoorthi
Looped transformers and Mixture-of-Experts (MoE) offer complementary routes to efficient scaling: recurrence increases computational depth at fixed parameters, while MoE sparsity e…
FML-bench: A Controlled Study of AI Research Agent Strategies from the Perspective of Search Dynamics
Qiran Zou, Hou Hei Lam, Wenhao Zhao +11
AI research agents accelerate ML research by automating hypothesis generation, experimentation, and empirical refinement. Existing agent strategies range from greedy hill-climbing…
Escaping the Cognitive Well: Efficient Competition Math with Off-the-Shelf Models
Xingyu Dang, Rohit Agarwal, Rodrigo Porto +3
In the past year, custom and unreleased math reasoning models reached gold medal performance on the International Mathematical Olympiad (IMO). Similar performance was then reported…
On the Impossibility of Retrain Equivalence in Machine Unlearning
Jiatong Yu, Yinghui He, Anirudh Goyal +1
Machine unlearning seeks to selectively remove the "influence" of specific training data on a model's outputs. The ideal goal is Retrain Equivalence--behavior identical to a model…
Metacognitive Reuse: Turning Recurring LLM Reasoning Into Concise Behaviors
Aniket Didolkar, Nicolas Ballas, Sanjeev Arora +1
Large language models (LLMs) now solve multi-step problems by emitting extended chains of thought. During the process, they often re-derive the same intermediate steps across probl…
Auto-Discovery-Bench: Diagnosing Structured State Tracking in Oracle-Guided Discovery
Tingting Chen, Beibei Lin, Srinivas Anumasa +5
Interactive discovery requires agents to maintain and update structured beliefs over many rounds of feedback. Before evaluating agents in noisy, open-ended scientific environments,…
Data for Mathematical Copilots: Better Ways of Presenting Proofs for Machine Learning
Simon Frieder, Jonas Bayer, Sam Looi +13
The datasets and benchmarks commonly used to train and evaluate the mathematical capabilities of AI-based mathematical copilots (primarily large language models) exhibit several sh…
-TCVAE: On the relationship between Disentanglement and Diversity
Cristian Meo, Louis Mahon, Anirudh Goyal +1
While disentangled representations have shown promise in generative modeling and representation learning, their downstream usefulness remains debated. Recent studies re-defined dis…
Masked Generative Priors Improve World Models Sequence Modelling Capabilities
Cristian Meo, Mircea Lica, Zarif Ikram +6
Deep Reinforcement Learning (RL) has become the leading approach for creating artificial agents in complex environments. Model-based approaches, which are RL methods with world mod…